• Journal of KIISE:Software and Applications
• /
• v.36 no.8
• /
• pp.633-641
• /
• 2009

In this paper, we proposed a new pattern classifier which can be incrementally learned, be added new class in learning time, and handle with analog data. Proposed pattern classifier has hierarchical structure and the classification rate is improved by using different metric for each levels. Proposed model is based on the Gaussian ARTMAP which is an artificial neural network model for the pattern classification. We hierarchically constructed the Gaussian ARTMAP and proposed the Principal Component Emphasis(P.C.E) method to be learned different features in each levels. And we defined new metric based on the P.C.E. P.C.E is a method that discards dimensions whose variation are small, that represents common attributes in the class. And remains dimensions whose variation are large. In the learning process, if input pattern is misclassified, P.C.E are performed and the modified pattern is learned in sub network. Experimental results indicate that Hierarchical Gaussian ARTMAP yield better classification result than the other pattern recognition algorithms on variable data set including real applicable problem.

• Journal of KIISE:Computing Practices and Letters
• /
• v.6 no.4
• /
• pp.430-440
• /
• 2000

To support Integrated Services in the Internet, the Internet Engineering Task Force(IETF) has developed new mechanisms. These new mechanisms include Resource Reservation Protocol(RSVP) and Integrated Services model. RSVP and Integrated Services are independent of the underlying link layer technologies and it is necessary to define the mapping of RSVP and Integrated Services specifications onto specific link layer technologies. IETF proposed a Subnet Bandwidth Manager(SBM) providing a method for mapping RSVP onto IEEE802 style networks. However, the SBM has several constraints and problems. For solving these constraints and problems, we design and implement a new Ethernet Bandwidth Manager(EBM). This paper discusses (1) new bandwidth management mechanism independent of RSVP, (2) simulation of Ethernet bandwidth to analyze the characteristics of Ethernet and accomplish Ethernet bandwidth management, (3) design and implemetation of EBM that allows Controlled-Load Service, and (4) results of our experiments on EBM.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.5
• /
• pp.101-107
• /
• 2009

In this paper, we describe a method of detecting emergency situation using images from surveillance cameras and propose a mobile robot tracking system for detailed examination of that situation. We are able to track a few persons and recognize their actions by an analyzing image sequences acquired from a fixed camera on all sides of buildings. When emergency situation is detected, a mobile robot moves and closely examines the place where the emergency is occurred. In order to recognize actions of a few persons using a sequence of images from surveillance cameras images, we need to track and manage a list of the regions which are regarded as human appearances. Interest regions are segmented from the background using MOG(Mixture of Gaussian) model and continuously tracked using appearance model in a single image. Then we construct a MHI(Motion History Image) for a tracked person using silhouette information of region blobs and model actions. Emergency situation is finally detected by applying these information to neural network. And we also implement mobile robot tracking technology using the distance between the person and a mobile robot.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.4
• /
• pp.46-63
• /
• 1995

This thesis described an unlimited vocabulary connected speech recognition system using Time Delay Neural Network(TDNN). The recognition unit is the diphone unit which includes the transition section of two phonemes, and the number of diphone unit is 329. The recognition processing of korean connected speech is composed by three part; the feature extraction section of the input speech signal, the diphone recognition processing and post-processing. In the feature extraction section, the extraction of diphone interval in input speech signal is carried and then the feature vectors of 16th filter-bank coefficients are calculated for each frame in the diphone interval. The diphone recognition processing is comprised by the three stage hierachical structure and is carried using 30 Time Delay Neural Networks. particularly, the structure of TDNN is changed so as to increase the recognition rate. The post-processing section, mis-recognized diphone strings are corrected using the probability of phoneme transition and the probability o phoneme confusion and then the eojeols (Korean word or phrase) are formed by combining the recognized diphones.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.25 no.6
• /
• pp.1435-1447
• /
• 2015

Finance Companies are operating a security solution such as E-DRM(Enterprise-Digital Right Management), Personal information search, DLP(Data Loss Prevention), Security of printed paper, Internet network separation system, Privacy monitoring system for privacy leakage prevention by insiders. However, privacy leakages are occurring continuously and it is difficult to the association analysis about relating to the company's internal and external distribution of private document. Because log system operated in the separate and independent security solutions. This paper propose a systematic chains that can correlatively analyze business systems and log among heterogeneous security solutions organically and consistently based on security documents. Also, we suggest methods of efficient detection for Life-Cycle management plan about security documents that are created in the personal computer or by individual through the business system and distribution channel tracking about security documents contained privacy.

• Journal of KIISE:Information Networking
• /
• v.31 no.4
• /
• pp.363-374
• /
• 2004

In this paper, we propose an in-line mode NIDS using network processors(NPs) that achieve performance comparable to ASIC and flexibility comparable to general-purpose processors. Even if many networking applications using NPs have been proposed, we cannot find any NP applications to NIDS in the literature. The proposed NIDS supports packet payload inspection detecting attacks, as well as packet filtering and traffic metering. In particular, we separate the filtering and metering functions from the complicated and time-consuming operations of the deep packet inspection function using two-level searching scheme, thus we can improve the performance, stability, and scalability of In-line mode system. We also implement a proto-type based on a PC platform and the Agere PayloadPlus (APP) 2.5G NP solution, and present a payload inspection algorithm to apply APP NP.

• Journal of Korean Society of Transportation
• /
• v.23 no.8 s.86
• /
• pp.203-213
• /
• 2005

Recently, with the increase of the interest to global environmental problems, bicycle has been recognized as the most environmentally friendly transportation mode. To promote cycling, it is necessary to ensure road space that bicycle can keep running safely and smoothly. This paper aims to evaluate the district with rearranged road space and network, given priority to bicycle, comparing the traditional district which are given priority to car from viewpoints of environment, safety and accessibility. As evaluation indicators, the carbon dioxide emission on the environment. the number of collision between car and bicycle on safety and an accumulated frequency measure on accessibility were used. As the result, it was clarified that bicycle road measures to create bicycle road by reallocation of road space and form bicycle exclusive network were effective.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.8
• /
• pp.915-919
• /
• 2014

Since smart phones are widely used globally, the demand for high-speed mobile communications are increasing. To accommodate the increased communication demand, the next-generation mobile communication services such as LTE-Advanced are deployed. In the LTE-Advanced base stations, the base station schemes of Digital Unit (DU) and Radio Unit (RU) are widely used. Here, the link between DU and RU is called fronthaul link. In the current fronthaul link, a digital optical communication interface is used, which is called Common Public Radio Interface (CPRI). However, the CPRI link cannot support the increased mobile traffic efficiently. Therefore, in this paper, we investigate an alternative technology based on the Radio over Fiber (RoF) to accommodate the increased mobile traffic economically.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.6
• /
• pp.5-14
• /
• 1994

This paper is about an experiment of speaker-independent automatic Korean spoken words recognition using Multi-Layered Perceptron and Error Back-propagation algorithm. The object words are 50 citynames of D.D.D local numbers. 43 of those are 2 syllables and the rest 7 are 3 syllables. The words were not segmented into syllables or phonemes, and some feature components extracted from the words in equal gap were applied to the neural network. That led independent result on the speech duration, and the PARCOR coefficients calculated from the frames using linear predictive analysis were employed as feature components. This paper tried to find out the optimum conditions through 4 differerent experiments which are comparison between total and pre-classified training, dependency of recognition rate on the number of frames and PAROCR order, recognition change due to the number of neurons in the hidden layer, and the comparison of the output pattern composition method of output neurons. As a result, the recognition rate of $89.6\%$ is obtaimed through the research.

• Journal of Korea Multimedia Society
• /
• v.2 no.1
• /
• pp.47-58
• /
• 1999

In this paper, we design the radio interface protocols of wideband Wireless Local Loop(WLL) that uses wideband Code Division Multiple Access(CDMA), and propose the random access procedures on access channel and analyze the performance of the channel. The structure of radio interface protocols are designed by three modular parts to meet the requirements of IMT-2000 : radio resource controlling part that depends on the radio resources, connection controlling part that is independent of the radio resources, and application part that supports the interface between different parts and user terminals. In addition, we propose the random access procedures on access channel and analyze the performance of the channel. We assume that both near-far effect and shadow fading can be compensated by open loop power control. We observe that the number of access probes(delay) depend on the initial mobile transmission power and power increment, and the poor estimation of initial mobile transmission power might increase the average number of access probes.